Some of our favorite development papers of 2018

Development Impact will now be on break over the next couple of weeks for the holidays, resuming in early January. As we did last year, we will turn to the holidays with a gift to you of some of our favorite development papers that we came across this year, mostly focusing on papers that we haven’t already blogged about.

Three from David McKenzie

• Flexible microfinance: Marianna Battaglia, Selim Gulesci and Andreas Madestam conduct an experiment with BRAC in Bangladesh, where they offer microfinance loan clients the opportunity to choose up to two times on an annual loan to postpone a month’s microfinance payment until the end of the loan. They find this added flexibility encourages women taking microfinance loans to undertake riskier investments and their businesses grow, while the lender does not experience any higher default. They do some good work on mechanisms, including eliciting ex ante demand uncertainty to show results are strongest for those facing uncertain demand.
• Border walls: Treb Allen, Caue Dobbin and Melanie Morten look at what happened when 548 miles of border fence were added or reinforced between 2007 and 2010. Obviously very topical, and very hard to measure given the state of migration data, the interconnectedness of markets, and the nature of the intervention – so lots of work on new data sources and structural spatial modelling. They find it cost $7 per person in the U.S., hurt high-skilled U.S. workers and Mexican workers, and only marginally benefited low-skill American workers.  Planet Money’s the Indicator podcast has a recent discussion.
• Low demand in female-dominated sectors: Morgan Hardy and Gisella Kagy have two papers (paper 1, paper 2) which use data and an experiment in Ghana to propose and provide evidence for a new explanation for why female business owners earn much lower profits than male owners in many developing countries –demand-constraints. The argument is that women crowd into a small number of sectors, so that there are few customers for any one of them, and that lack of demand makes them unprofitable. I like this paper for both suggesting a new explanation for gender gaps to me, as well as for all the questions it raises about when this could be true, and what the implications for policy interventions are.

Three from Dave Evans
I’ve been presenting some new work on interventions to improve girls’ access to and learning in school, and one of the most consistent reactions is that schools can do much more than just help kids learn academic skills (although they should definitely do that). Two of my favorite papers show how school-based programs can do more.

• Changing gender norms at school: New work by Dhar, Jain, and Jayachandran shows that a school-based training to improve gender equity in India led to more gender equitable beliefs and also behavior, with boys doing more household chores. The authors also demonstrate a way to check the robustness of results to social desirability bias (when people give the answers that they think are most socially acceptable). 
• Teaching negotiation skills after school: Ashraf et al. show that providing negotiation training for girls in Zambia helped them advocate for themselves in their households, including more time for schoolwork at home and ultimately improved educational outcomes. A lot of organizations are providing various forms of life skills training: Here’s one with a measurable impact.

My third paper highlights one of the great challenges: moving from a successful pilot to effective implementation at scale.

• Improving reading at scale: In Kenya, the government has recently implemented the “Tusome national literacy program.” It builds on a series of pilots that were closely evaluated. The government incorporated many of the results of the pilot evaluations into the scaled intervention. This work by Piper et al. demonstrates that it’s possible to achieve gains to literacy at scale through a multi-faceted intervention that includes structured teacher guides, coaching for teachers, student textbooks, and more. 

Four from Berk Ozler
Family Planning:

• Holt, Dehlendorf, and Langer (Contraception, 2017) on quality contraceptive counseling: For a long time, the prevailing paradigm in counseling clients on contraceptive methods has been an informed choice model in which individuals are given extensive information in order to make their own independent choices. But, the field is recognizing that this is an unrealistic model to achieve informed choice in many settings and has been moving towards a shared decision-making paradigm. How to do this while treating the clients with respect and empathy, earning their trust, and protecting their privacy is the subject of this paper on a revised framework on how to provide quality counseling to improve both patient experience and public health.

Sex work:

• Manian (WP, 2018) on health certification for sex workers in Senegal: Shanti Manian investigates why health certification rates remain low among sex workers in Senegal, when theory would predict that, in the presence of asymmetrical information, certificate should provide a premium to the worker. They find that client demand for the certificate is low, likely because the certification does not provide much additional information about the STI status of the provider. Furthermore, confidentiality concerns and stigma play a role in the reported unwillingness of sex workers to get certified at any incentive amount. The point about the potentially large role self-stigmatization plays (unwillingness to accept a “sex worker” identity, i.e. admitting only to self) is novel and important.

Cash transfers:

• Kasy (WP, preliminary, 2018) on UBI vs. EITC: Max Kasy makes two economic and two moral/political arguments in favor of a UBI over subsidies of low-wage work, such as the EITC. I can’t put my finger as to why this paper stuck in my head, but the fact that it did – despite the evidence on the success of EITC as an anti-poverty tool in the U.S. (see here and here) – is perhaps the reason to read this short piece. Kasy is also doing other interesting work – whether it is randomization in RCTs, publication bias, etc., so may be worth following…
• Honorable mention. Papers on longer-term effects of cash transfers: Barham, Macours, and Maluccio (WPs, 2018a & 2018b); Blattman, Fiala, and Martinez (WP, 2018)

Taxation:

• Okunogbe and Pouliquen (WP, 2018) on technology, taxation, and corruption: Our colleague Oyebola, with her co-author Victor, investigates the heterogeneous effects of e-filing among small and medium firms in Tajikistan and finds that the distribution of tax payments changes in a way that is arguably more equitable.

 
A final bonus: Some highlights of impact evaluation work published by some of our younger (defined as “Economist” or equivalent to assistant professor) colleagues this year

• Sam Asher and Paul Novosad have a paper forthcoming in the AER which examines the impact of rural roads on economic development in India, using fuzzy RD and massive amounts of census data.
• Maria Jones and Florence Kondylis have a paper in the JDE which uses an experiment to test how eliciting feedback affects the demand for agricultural services in Rwanda.
• Eeshani Kandpal and co-authors have a paper forthcoming in the JHR which uses a randomized evaluation of a CCT in the Philippines to examine impacts on child labor.
• Martin Kanz and co-authors have a couple of papers on experiments with credit cards in Indonesia. The first, published in the QJE, looks at the use of platinum cards as status symbols; the second, forthcoming in the JPE, looks at how moral incentives to repay cards work for an Islamic Bank.
• Owen Ozier has a paper in the AEJ Applied that looks at the long-run impacts of deworming, using positive externalities from an experiment in Kenya; and a paper in the JHR that uses an RD design to look at the impacts of secondary schooling in Kenya.
• Caio Piza has a paper in the JDE that uses RD and DiD to re-examine the impact of a simplification program on formalization in Brazil.
• Gil Shapira and co-authors have a paper in Health Economics that reports on two randomized experiments that test demand-side and supply-side incentives for improving health care in Rwanda.